Each row is one line of the log that imports write as they run: one row for each row processed, imported or not. When a contact was imported but list subscriptions were skipped, there is one row per skipped subscription instead (res 2). Imports started from the app's import tool, imports submitted through POST /import/bulk_import, older sync imports and integrations' automatic syncs all write rows here. This endpoint returns every row in the account, with no way to narrow it to one import: filters and orders are ignored, and rows have no defined order. To see how one import went, use the endpoint for its kind: for an import started from the app or by an integration, GET /import/history to find it (for an app import, its processid matches the processid on these rows) and GET /import/history/{importId}/{type} for its failed or partly imported rows and their reasons; for an import submitted through POST /import/bulk_import, GET /import/info with its batchId. Use this endpoint only to fetch specific rows by ids. Paginated with limit (default 20, at most 100) and offset. No permission beyond API access is checked: any API user can read every row in the account, including emails from other groups' imports, unlike GET /import/history.